Exchange traded fund providers have been slashing fees to dirt cheap levels. While cheap investments are an attractive draw, people should not solely invest in a fund based on low expense ratios.

Last month, BlackRock’s iShares reduced the fees on a number of its products, with the Shares Core S&P Total US Stock Market ETF (NYSEArca: ITOT) showing an expense ratio of 0.03%. ITOT was the cheapest on the block for a brief moment. [The New Cheapest ETF In The U.S. Is A Familiar Face]

Not to be outdone, Charles Schwab lowered fees by one basis point on four of its large-cap ETFs in response, with the Schwab U.S. Large-Cap ETF (NYSEArca: SCHX) and Schwab U.S. Broad Market ETF (NYSEArca: SCHB) both coming in at a low 0.03% expense ratio. [Schwab Responds to iShares Fee Cuts]

The cheap expense ratios are great for investors but fees should not solely dictate investment decisions.

“I’m glad fees are coming down, but I maintain that a 2-basis-point fee difference is a lousy way to pick a fund,” Paul Britt, senior analyst at FactSet Research Systems, told Eric Rosenbaum for CNBC.

The cost of ETF investing is more than just a couple of basis points in the expense ratio. While low expense ratios are nice for keeping costs down in long-term buy-and-hold investments, investors should also consider other implicit costs. [How to Pick ETFs as Fee War Heats Up]

For instance, investors should consider tracking error, or how much an ETF’s return may diverge from the underlying benchmark index. A number of factors may contribute to the tracking divergence, including the wide market bid/ask spreads, portfolio optimization, internal trading costs, portfolio management styles and securities lending fees. These factors would help contribute to an ETF’s price divergence from its net asset value.